Sign inGet started
Skills / Productivity-habits

OKR Framework

An Objective is simply WHAT is to be achieved, no more and no less. Key Results benchmark and monitor HOW we get to the objective. Effective KRs are specific and time-bound, aggressive yet realistic. Most of all, they are measurable and verifiable.

John Doerr's OKR (Objectives and Key Results) framework is the goal-setting system behind Google, Intel, and the Gates Foundation. It separates the inspirational 'what' (Objectives) from the measurable 'how you'll know' (Key Results), creates bidirectional alignment from CEO to individual contributor, and operates on a 0.0–1.0 scoring rubric that makes success undeniable. This skill applies Doerr's specific rules — not generic goal advice — to draft, audit, classify, and cascade OKRs across any level of an organization.

By John Doerr · Free
Specimen 01 · Live diagnosisOKR Framework
Input

“It's for my engineering team, quarterly. Our big focus is improving API reliability, but honestly when I write it out it always sounds vague.”

Diagnosis
Let's build your OKRs the way Doerr defines them — not just goals, but a set of Objectives that inspire and Key Results that prove success.
Full transcript ↓
Calibrated referenceagent-skills.ai
The gap

Set inspiring objectives, define quantified key results, and finally measure what actually matters

Doerr's OKR framework organizes goal-setting around two tightly defined components: Objectives (qualitative, inspirational, time-bound statements of what you want to achieve) and Key Results (3–5 quantified, outcome-based measures per objective that prove whether the objective was reached). The framework runs on dual cadences — annual OKRs for strategic direction and quarterly OKRs for operational execution — and requires all OKRs to be public and visible organization-wide to enable cascading alignment from company to team to individual. A critical distinction separates Committed OKRs (must be fully achieved, scored 1.0) from Aspirational/Stretch OKRs (moonshots where 0.7 is considered success). Doerr names four organizational benefits he calls the 'Four Superpowers': Focus and Commit to Priorities, Align and Connect for Teamwork, Track for Accountability, and Stretch for Amazing. The companion CFR system (Conversations, Feedback, Recognition) replaces annual performance reviews with continuous OKR-linked dialogue, and OKRs are explicitly separated from compensation to eliminate sandbagging.

The problem

Most goal-setting fails because objectives are vague ("improve customer experience") and key results are actually tasks ("run 3 customer surveys"). Without Doerr's strict separation of inspirational objectives from quantified outcome metrics, teams measure effort instead of impact — and never really know if they succeeded. Annual goal cycles get forgotten by Q2 because there's no scoring rubric, no transparency, and no alignment between what leadership wants and what individuals are working on.

The solution

Apply Doerr's precise OKR rules — not watered-down goal advice — to produce a quarterly OKR document where every key result is a measurable outcome with a defined target, every objective is genuinely inspiring, and the 0.0–1.0 scoring rubric makes success or failure unambiguous.

You bring
  • A goal, initiative, or area of focus you're trying to drive results in
  • Your role and organizational level (individual, team, company)
  • Existing OKRs you want audited or improved (optional)
  • The time horizon — quarterly or annual
You get
  • A complete, scored OKR set with 3–5 objectives and 3–5 key results each, ready to publish
  • Classification of each OKR as Committed (score target: 1.0) or Aspirational (score target: 0.7)
  • An audit report identifying anti-patterns: activity-based KRs, vanity metrics, goal bloat, or compensation linkage
  • A cascade map showing how your team or individual OKRs connect to higher-level company objectives
In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01The same situation, two ways
Specimens A · B
Specimen AVague
Unaided

“You're heading into Q2 planning with a slide that says 'Goals: improve reliability, ship new features faster, reduce churn.' Your team nods in the meeting but nobody's sure what success actually looks like, and by week six everyone is measuring different things — one engineer tracks deployments, another tracks uptime, the PM tracks NPS. At the quarter-end retro, you can't agree whether Q1 was good or not.”

×Vague — nothing concrete to act on.
Specimen BCalibrated
With OKR Framework

“You open Q2 with a public OKR document: three Objectives, each with four quantified Key Results, each classified as Committed or Aspirational with explicit score targets. At the six-week check-in your team scores each KR on the 0.0–1.0 scale — KR2 is at 0.4, which is a red flag under Doerr's rubric, and you reallocate two engineers before it's too late. At the quarter-end retro, there's no debate: you hit 0.9 on the aspirational KRs and 1.0 on the committed ones.”

Specific, grounded, and ends with one clear next step.
Fig.02A real session, unedited
In-app · assistant
OKR FrameworkClaude
OKR Framework
Let's build your OKRs the way Doerr defines them — not just goals, but a set of Objectives that inspire and Key Results that prove success. Before we draft anything, tell me: what level are we working at (individual, team, or company) and what's the time horizon — this quarter, or annual?
It's for my engineering team, quarterly. Our big focus is improving API reliability, but honestly when I write it out it always sounds vague.
Good — and 'vague' is exactly the right diagnosis. Here's the Doerr test: if your Objective sounds like it could already be true or could mean different things to different people, it fails. Try this as a starting Objective: 'Make our API a foundation our customers can bet their business on.' That's qualitative, time-bound to Q2, and impossible to misread. Now let's turn 'improve API reliability' into Key Results — which means metrics, not actions. What does success actually look like in numbers? Think: uptime, latency, incident frequency, time-to-recovery.
OK so something like: achieve 99.9% uptime, reduce P1 incidents from 8 to 2 per month, and get mean time to recovery under 15 minutes?
Those are excellent Key Results — all three pass the core test: specific target, measurable outcome, not a task. Now I need to classify each one. Uptime at 99.9% — is that an operational floor you must hit, or a stretch goal? If your current uptime is 99.5%, that's a Committed OKR: you need to achieve it fully (score 1.0). But if you're at 99.0% and 99.9% requires an architectural overhaul, that's Aspirational — Doerr would say a 0.7 score (99.6%) is still success. Getting the classification wrong is one of the most damaging OKR mistakes: it either burns out your team chasing a moonshot as if it's required, or lets them sandbag on something critical.
We're currently at 99.6% so 99.9% is genuinely a stretch. The incident reduction is more of a committed goal — we know exactly what's causing them.
Perfect — that's the right judgment call. So your OKR is: Objective (Aspirational): Make our API a foundation our customers can bet their business on. KR1 (Aspirational, target 0.7): Achieve 99.9% API uptime (current: 99.6%). KR2 (Committed, target 1.0): Reduce P1 incidents from 8 to 2 per month. KR3 (Committed, target 1.0): Reduce mean time to recovery to under 15 minutes. One question before we finalize: does this OKR connect to a company-level objective? Doerr's framework requires you to be able to trace each team OKR upward. If the company has an objective around customer trust or platform stability, your team OKR should visibly cascade from it.
Yes — the company OKR is 'Become the most trusted developer platform in our category.' This maps directly to that.
That's clean cascade alignment — exactly what Doerr describes. You can document it as: Team OKR supports Company Objective: 'Become the most trusted developer platform in our category.' Your quarterly OKR document is complete and publication-ready. Want me to draft the full formatted version, or audit whether you have any other OKRs this quarter that might be stepping on each other or exceeding the 3–5 objective limit Doerr recommends?
Reply to OKR Framework
What the skill is doing
01Objective Drafting
02Key Result Conversion
03Committed vs. Stretch Classification
Fig.03The same skill, in the channel you already use
WhatsApp · live
9:41
OFOKR Frameworktyping…
Messages are end-to-end encrypted. No one outside of this chat can read them.
Today
It's for my engineering team, quarterly. Our big focus is improving API reliability, but honestly when I write it out it always sounds vague.9:14✓✓
Let's build your OKRs the way Doerr defines them — not just goals, but a set of Objectives that inspire and Key Results that prove…9:14
Before we draft anything, tell me: what level are we working at (individual, team, or company) and what's the time horizon — this…9:15
Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

Reads the situation, names the pattern, returns one concrete next move.
Delivered in seconds, inside a conversation that already exists.
Specimen · WhatsApp Business API · live
Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Objective Drafting

Transforms a vague goal statement into a proper Objective: qualitative, inspirational, action-oriented, and anchored to a time horizon. The skill tests each Objective against Doerr's criteria — it must be significant enough to matter, concrete enough to focus effort, and motivating enough that a team would want to pursue it.

Based on Doerr's definition: Objectives are WHAT you want to achieve — not how, not when in detail, not by how much. They must pass the 'would this inspire my team?' test.
CapabilityC-02

Key Result Conversion

Converts activities, tasks, and vague intentions into proper Key Results: specific, quantified, outcome-based, and falsifiable. Crucially rejects any KR that describes an action rather than an outcome ("launch the feature" becomes "feature adopted by 40% of active users within 90 days").

Applies Doerr's hard rule that Key Results must be metrics with a clear target value — not to-do items. The test: if it doesn't have a number, it's not a Key Result.
CapabilityC-03

Committed vs. Stretch Classification

Determines whether each OKR should be classified as Committed (achievable with current resources, must score 1.0) or Aspirational/Stretch (a moonshot requiring breakthrough, where 0.7 is considered success). Misclassification is one of the most common OKR errors — treating stretch goals as committed leads to burnout; treating committed goals as aspirational enables sandbagging.

Directly from Doerr's two-category model: committed OKRs are operational necessities, stretch OKRs are growth bets. The scoring interpretation differs fundamentally between the two types.
CapabilityC-04

OKR Audit

Evaluates an existing set of OKRs against Doerr's full criteria and returns a scored assessment with specific remediation. Detects the most common failure modes: output-based KRs masquerading as outcomes, objectives that are actually strategies, too many OKRs diluting focus, and key results that can't be measured until after the quarter ends.

Applies Doerr's complete validity checklist, the Four Superpowers lens, and the 3–5 OKR / 3–5 KR structure guidelines from 'Measure What Matters.'
CapabilityC-05

Cascade Alignment

Maps a team's or individual's OKRs to the higher-level Objectives they support, ensuring the 60/40 bidirectional balance Doerr prescribes — where roughly 60% of OKRs originate bottom-up from employees while 40% cascade from leadership priorities. Surfaces misalignment where team work doesn't connect to any company objective.

Implements Doerr's bidirectional goal-setting model and the cascading alignment principle: company OKRs inform department OKRs, which inform team OKRs, which inform individual OKRs — but employees set the majority of their own.
Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces
OutputD-01

Quarterly OKR Document

A complete, publishable OKR set for a role, team, or company — 3–5 Objectives each with 3–5 Key Results, classified as Committed or Aspirational, with scoring targets and a cadence note. Formatted for direct use in a team meeting, board deck, or OKR tracking tool.

OutputD-02

OKR Audit Scorecard

A line-by-line evaluation of existing OKRs against Doerr's criteria, with each Objective and Key Result rated on validity, measurability, and classification accuracy. Includes a prioritized list of anti-patterns found and specific rewrites for each failing element.

OutputD-03

Cascade Alignment Map

A visual mapping showing how each team or individual OKR traces upward to a department or company Objective. Highlights orphaned OKRs (effort with no strategic connection) and over-concentrated alignment (all eggs in one company bucket).

OutputD-04

CFR Check-in Template

A structured quarterly conversation guide built on Doerr's CFR (Conversations, Feedback, Recognition) companion system. Gives managers a concrete agenda for OKR-linked check-ins that replace annual reviews with continuous, forward-looking dialogue.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

John Doerr

John Doerr is a venture capitalist at Kleiner Perkins who learned OKRs directly from Andy Grove at Intel in the 1970s and personally introduced the framework to Google in 1999 — when the company was 40 people working out of a garage. His 2018 book 'Measure What Matters' became a New York Times bestseller and documents OKR adoption at Google, Intel, the Gates Foundation, Bono's ONE Campaign, and Adobe. His TED Talk on OKRs has been viewed over 6 million times.

Status · Inspired by John Doerr’s work — not yet claimed. Are you John Doerr?
Primary sourceS-01

Measure What Matters

by John Doerr

Partner at Kleiner Perkins; early Google board member; introduced OKRs to Google in 1999; NYT bestselling author of 'Measure What Matters'; TED Talk with 6M+ views.

Read the original ↗
Citationwhatmatters.com
In the build queue

Be first to run it.

OKR Framework is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail
At launchI need to write OKRs for my team this quarter. We've done OKRs before but they always end up looking like a to-do list — can you help me figure out what's wrong and rebuild them the right way?